AI Analysis May Improve Vaping Cessation Efforts on Social Media

Social media platforms are widely used to promote e-cigarette products as safer alternatives to traditional tobacco. High school students are often exposed to these ads and messages, leading to higher rates of use. While regulatory policies such as the Food and Drug Administration’s moratorium on approving flavored vapes and the Tobacco 21 legislation are aimed at curbing e-cigarette use among young people, researchers at the University of Rochester see opportunities to promote effective vaping cessation on social media platforms.

Using AI to Analyze Vaping Social Media Posts

In their recently awarded National Institutes of Health (NIH) Research Project Grant Program, “Artificial Intelligence for effective communication to promote vaping cessation on social media,” UR CTSI researchers Dongmei Li, PhD, Zidian Xie, PhD, and Karen Wilson, MD, MPH, working with Chenliang Xu, PhD, associate professor of Computer Science, and Ana Paula Cupertino, PhD, will use natural language processing techniques and deep learning models to identify the key features of high-engagement social media posts related to e-cigarette products. They will then deploy these techniques to collect data from Twitter/X, Instagram, TikTok, and YouTube, where vaping and e-cigarette posts attract high user engagement.

“Our current project focuses on identifying and validating key features in vaping-related social media posts associated with high user engagement,” Li said. “These features will provide valuable guidelines for designing future vaping prevention messaging campaigns with high user engagement to maximize their dissemination to the public.”

A follow-on project will leverage AI technology to develop and validate vaping prevention and cessation posts for multimedia campaigns. These posts will utilize text, image, and video and target young people and disadvantaged groups, specifically.

Perception Affects Behavior

According to Li, the biopsychosocial model of addiction suggests that tobacco product use may be affected by users’ perception of risk associated with those products. The researchers believe their work can be used to increase the public perception of e-cigarette use as a risky behavior and foster a reduction in use.

“Studies have shown that exposure to vaping promotion on social media is associated with the initiation of vaping, while vaping prevention content exposure is associated with reduced vaping,” Li said.

Zidian Xie, PhD, is a research data engineer II in the Informatics and Analytics branch at UR CTSI. His research focuses on applying AI technologies such as natural language processing and large language models to understand and prevent substance use on social media.

“The vast and valuable data within social media posts offers crucial insights into public engagement,” Xie said. “Advances in AI technology, particularly large language models, empower us to efficiently analyze thousands, and even millions, of these posts.”

Current vaping prevention campaigns have seen low impact because of their lack of engagement.

“Results from the proposed study will provide valuable guidance in designing effective vaping prevention messages for future public health campaigns to help with the effective communication of risks associated with e-cigarette use, which will help prevent the initiation and counter-uptake of e-cigarettes by youth and young adults,” Li said.

Research reported in this publication was supported by the National Cancer Institute of the National Institutes of Health under Award Number R01CA285482. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Jonathan Raab | 9/10/2024